880 Brief Communication
Categorizing sex and identity from the biological motion
of faces
H. Hill and A. Johnston
Head and facial movements can provide valuable
cues to identity in addition to their primary roles
in communicating speech and expression [18]. Here
we report experiments in which we have used
recent motion capture and animation techniques to
animate an average head [9]. These techniques
have allowed the isolation of motion from other cues
and have enabled us to separate rigid translations
and rotations of the head from nonrigid facial
motion. In particular, we tested whether human
observers can judge sex and identity on the basis of
this information. Results show that people can
discriminate both between individuals and between
males and females from motion-based information
alone. Rigid head movements appear particularly
useful for categorization on the basis of identity,
while nonrigid motion is more useful for
categorization on the basis of sex. Accuracy for
both sex and identity judgements is reduced when
faces are presented upside down, and this finding
shows that performance is not based on low-level
motion cues alone and suggests that the
information is represented in an object-based
motion-encoding system specialized for upright
faces. Playing animations backward also reduced
performance for sex judgements and emphasized
the importance of direction specificity in admitting
access to stored representations of characteristic
male and female movements.
Address: Department of Psychology, University College London,
Gower Street, London WC1E 6BT, United Kingdom.
Correspondence: H. Hill
E-mail: harold.hill@ucl.ac.uk
Received: 7 February 2001
Revised: 9 March 2001
Accepted: 18 April 2001
Published: 5 June 2001
Current Biology 2001, 11:880885
0960-9822/01/$ see front matter
2001 Elsevier Science Ltd. All rights reserved.
Results and discussion
The fact that impersonators can mimic the ways in which
famous people move their heads and faces demonstrates
that, in addition to the primary role they have in communi-
cation, these movements can provide cues to sex and
identity independently of the underlying shape and tex-
ture of the face. Similarly, in computer-animated films, a
characters expressions and voice can be derived from an
actor, as in Tom Hanks performance as Woody in Pixars
Toy Story [10]. The characters face and head movements
mimic those of the actor even though their underlying
shapes are quite different. We report experiments in
which we computer-animated an average head with move-
ments captured from real people in order to investigate
whether motion provides useful information for categoriz-
ing faces.
The animation process is illustrated and described in Fig-
ure 1. Four different movement sequences were captured
for each of twelve actors (we use the term as a short-
handthe volunteers were not trained actors) and were
used to animate the same three-dimensional model of an
average face [9]. Each animation was of a person telling
a two-line question-and-answer joke to another individual
(e.g., Why do cows have bells? Because their horns dont
work!). This activity was intended to elicit expressive
and natural facial gestures, expressions, and speech from
the actor.
All the stimuli produced in this manner were physically
identical at the start of each animation and differed only
in the way that they moved. This allowed us to investigate
motion-based information independently of other cues.
The technique also allowed the separation of rigid head
motion, in which the head translates and rotates but does
not change shape, and nonrigid motion, in which the
expression changes but the head does not move. Given
their different natures, these two components may be
processed independently and may make different contri-
butions to the perception of face-based biological motion.
By concentrating on motion, we do not intend to deny
the importance of other sources of information for these
tasks. Everyday experience and numerous experiments
show that we can recognize sex and identity from static
photographs that provide no motion information. How-
ever, motion is fundamental to vision, and diagnostic dif-
ferences may be used by a system that makes use of any
and all available discriminating information. The available
evidence suggests motion may be particularly useful when
spatial and other cues are degraded or changed, for exam-
ple by presentation in photographic negative [5, 6]. In
previous studies, researchers have used moving-point
light stimuli to look at biological motion in general [11, 12]
and facial movement in particular [1, 2], but the current
technique provides more natural motion information
 Brief Communication 881

Figure 1

Examples of the animations used as stimuli

(please also see the Supplementary
material). The movement of the faces was
captured by a pair of digital video cameras.
The top row shows the original movement
sequence from the right hand camera, the
middle row shows an animation from the same
viewpoint, and the bottom row shows the
animation as the observers viewed it. The
middle row was not used in any of the
experiments, but we include it to facilitate
comparison between the original and the
animation. In each case, the leftmost image
shows the neutral starting position. The
following frames were taken at 1 s intervals.
This sequence was 6.7 s long, excluding pre-
and post-masks. The average length of
sequences was 7.2 s (standard deviation,
1.6 s). The motion of the 17 markers and the
pupils were automatically tracked with
Famous vTracker (Famous Technologies) from the movements of the marker. The markers on rendered with 3D Studio Max (Kinetix) for the
video footage taken from the two cameras the forehead, temples and nose were used production of 640 480 pixel 25-frames-
placed approximately 15 either side of the for defining rigid translations and rotations of per-second avi format movies. Animations
direction in which the actors were facing and the head. Because rigid motion can be fully were compressed with Radius cinepak, with
at a distance of 1 m. The cameras were characterized by the motion of a few markers 90% compression quality and a keyframe
calibrated for each recording with a in a way that nonrigid motion cannot, the rigid every 15 frames. A 10 frame mid-gray mask
calibration object of known dimensions so that component of the animations was more was added to the beginning and end of each
2D tracked positions could be converted into accurate. In both cases the timing of animation. Backward and inverted
3D positions on the basis of projective movement is accurately captured given 25 animations were rendered with the same
geometry. These were used for the animation frames per second temporal resolution, parameters as for upright forward animations,
of a three-dimensional head model produced whereas spatial properties cannot be truly but with the order or orientation of the images
as an average of 200 heads, 100 male and veridical given the differences in shape altered. In all experiments, observers could
100 female [9], with the number of vertices between actor and average head model as view the animations as many times as they
reduced to 65,525 for import into 3D Studio well as the limited spatial sampling. However, wished. Presentation was controlled by
Max. Animation was accomplished with the these limitations were constant for all Microsoft Mediaviewer, and responses were
commercially available Famous animator animations and experimental conditions and recorded manually in the rating and sorting
(Famous technologies). In this system, marker so should not have biased the results. experiments and by programs written with
positions are associated with a hotspot and Animated head models were texture- Macromedia Director for the 2-AFC and odd-
an area of influence on the model that inherit mapped with an average texture [9] and one-out tasks.

while more fully eliminating residual spatial cues, such
as the aspect ratio of the underlying face.
There are a number of ways in which motion may be
useful for distinguishing between faces. It can provide
indirect cues to three-dimensional shape via structure-
from-motion [13]. However, mathematical analysis of this
process assumes rigid motion [13], an assumption satisfied
by head movements but violated by most facial move-
ments. In previous experiments [18] the cues motion
provides about shape may have been important, but in
the present experiments any such cues would be limited
because the underlying shape was always the same. How-
ever, differences in shape do result as a consequence of
differences in movement, and the resulting differences
in shape may provide useful information. This highlights
the difficulty of completely separating motion from spatial
information. Also, differences in the ways that people
move their faces may result in useful and reliable differ-
ences in low-level image motion. Some people may move
the whole or parts of their faces more than others. Lastly,
the spatial changes occurring over time for a moving object
produce a spatio-temporal signature that in itself may
be useful for recognition [14].
In order to investigate whether the biological motion of
faces provides cues to identity, we used two tasks. In
the first, observers were presented with 16 animations, 4
different animations for each of 4 actors, and were asked
to sort these into 4 equally sized groups on the basis of
identity. Observers could view the animations in any order
as many times as they wished, and they sorted the anima-
tions simply by moving their icons into groups on the
screen. All the stimuli used in this experiment were gener-
ated with movement sequences captured from male
actors. This method ensured that the ability to do this
task was independent of any ability to categorize sex.
Different groups of observers saw rigid head motion (N
15), nonrigid facial motion (N 16), or combined rigid
head and nonrigid facial motion (N 16). The observers,
like the actors, were recruited from the student population
of University College London. The task, like recognition,
882 Current Biology Vol 11 No 11
Figure 2
The results of the identity sorting experiment, in which four different
animations of four actors were sorted into equally sized groups
on the basis of identity. Different groups of observers saw nonrigid
facial motion alone (N 16), rigid head motion alone (N 15), or
both types of motion combined (N 16). Performance was scored
according to how many other examples of the same actor were put
in the same group for each animation. This gives a maximum score of
48 (4 actors 4 examples 3 other examples in the group). The
minimum score, when each group contains only one example of each
actor, is 0. Chance, calculated by the Monte Carlo method of
generating and averaging the score for 10,000 of the 16! possible
ways to sort 16 animations, was 9.6. We also tested whether average
scores for our groups of observers were significantly different from
chance by calculating the proportion of times that the observed
scores were exceeded by random samples of the same size. Combined
rigid and nonrigid and rigid alone stimuli were both sorted
significantly better than chance would have allowed (p .05), and
nonrigid stimuli alone were sorted marginally better (p .1). Error
bars show standard errors.
required that the description of the motion that observers
recovered was stable enough to generalize over different
examples of the same face while sensitive enough to dis-
tinguish between examples of different faces [15]. This
task allowed us to investigate the motion information
essential for recognition independently of memory-based
or cue conflict effects that would be involved in recogni-
tion per se.
Performance in this task was scored for each animation
according to how many other examples of the same person
were sorted into the same group (Maximum score 48,
minimum 0, chance 9.6 by Monte Carlo simulation).
Results are summarized in Figure 2, and details of the
scoring system are given in the legend. In order to avoid
having to make assumptions about the scoring distribu-
tion, we used Monte Carlo and Bootstrap resampling sta-
tistical methods to analyze the data [16]. There was a
significant effect of type of motion shown (p .05), with
performance significantly above chance for rigid motion
and for combined rigid and nonrigid motion (p .05).
Performance was marginally above chance for nonrigid
motion (p .1). Rigid head motion appears particularly
useful for categorizing people on the basis of identity.
The difference between performance in the rigid head
motion alone and nonrigid facial motion alone conditions
was significant (p .05), and the other two differences
between conditions were marginally significant (p .1).
Nonrigid facial motion is less useful than rigid motion
for characterizing individuals. It is possible that nonrigid
facial movement could interfere with identity judgements
in this task given that the facial speech of two people
telling the same joke or making the same expression may
be more similar in many ways than that of the same person
telling different jokes or making different expressions.
Although this factor was not fully balanced, two jokes
were told by all of the actors used in this experiment, so
in order to test whether subjects grouped faces on the
basis of the joke told, we scored the data according to
whether examples of different actors telling these jokes
were grouped together. In this case, sorting scores were
no different than what would have been due to chance.
This result shows that observers were not sorting on the
basis of the joke told and validates the sorting task and
scoring method.
We also used an odd-one-out task to test whether motion
provides cues to identity. Twelve naive observers were
presented with seventy-two trials each consisting of three
animationstwo different examples of one person and
one example of a different person of the same sex. The
observers task was to identify the animation derived from
the unique individualthe odd one out. Observers initi-
ated presentation of the stimuli and responded by using
an application written in Macromedia Director. In order
to provide clues as to the critical properties of the motion
information used for this task, we compared inverted and
backward play to normal (forward and upright) presenta-
tion for stimulus triplets. Inverted presentation leaves
low-level motion cues the same but is well known to
adversely affect many aspects of face processing [17]. Play-
ing an animation backward uses the same static frames
as the same animation played forward but changes the
overall pattern of movement. Both inverted and backward
play test the extent to which performance can be achieved
on the basis of perceptual matching or whether stored
knowledge is needed, as they leave the perceptual similar-
ities available for matching the same but might be ex-
pected to affect access to stored knowledge about how
faces normally appear. All the stimuli used in this experi-
ment contained both rigid head and nonrigid facial
motion.
As can be seen from the results summarized in Figure 3,
the accuracy with which the odd one out was identified
depended upon how the stimuli were presented. A one-
way repeated-measures ANOVA on the proportion of cor-
rect responses showed a main effect of the presentation

Figure 3
The results of the odd-one-out experiment in which observers (N
12) had to choose which animation corresponded to the unique
individual from a choice of three. The other two animations were two
different examples of another individual. Performance was above
chance in all conditions but significantly worse when animations were
shown inverted. Error bars show standard errors.
condition [F(2,22) 8.7, p .05]. Post hoc paired
t tests showed that playing inverted stimuli produced
significantly worse performance than normal [t(11) 4.4,
p .05] or backward [t(11) 4.4, p .05] presenta-
tion. Normal and backward presentation did not differ
from each other (p .1). One-sample t tests showed
that performance was significantly above chance (33%) in
all conditions [normal: t(11) 8.9, p .05; backward:
t(11) 7.7, p .05; and inverted: t(11) 5.7, p .05].
The detrimental effect of inversion shows that perfor-
mance is not based upon low-level properties of motion
alone (for example, gross amount of head motion), as this
information would be recoverable as easily from inverted
stimuli. Instead, it appears that identity specific-motion
information is processed by a system tuned to upright
faces. Backward play did not affect performance, and this
result shows that even when played backward, motion
contains information that allows us to discriminate be-
tween individuals. Either discriminating cues are static,
direction independent, and/or temporally symmetric, or
playing animations backward generates new but equally
discriminable patterns of movement. Previous evidence
showing that we are less good at recognizing faces that
have been learned normally from videos played backward
[7] favors the latter explanation.
To extend the evidence obtained from the identity-based
tasks used so far, we also tested whether observers could
recover information about the sex of the actors from these
animations. No training was given, so any ability to do
this depended both on there being differences between
the ways males and females move their faces and on
observers being able to extract these cues from the anima-
Brief Communication 883
Figure 4
The results of an experiment in which observers rated the sex of
animations on a 6 point scale with 1 indicating definitely male and
6 indicating definitely female. The same observers took part in the
same conditions as described for Figure 2. Ratings for male and
female items were significantly different, but the amount of difference
depended on the type of movement shown. Rigid head movements alone
appear least useful for discriminating sex. Error bars show standard
errors.
tions and relate them to their knowledge of sex differ-
ences. Sex judgements cannot be achieved on the basis
of perceptual matching alone, as they require access to
stored knowledge.
In the first sex judgement experiment, 48 observers rated
the sex of all 48 animations on a scale of 1 to 6, with 1
indicating definitely male and 6 definitely female (or vice
versa for half the observers). Observers controlled presen-
tation of the stimuli by using Microsoft Mediaviewer and
responded by hand on a prepared ratings form. Three
groups, each with a different set of 16 observers, saw
nonrigid facial motion, rigid head motion, or combined
rigid head and nonrigid facial motion (the observers were
the same as those who subsequently sorted the stimuli
according to identity, as reported above).
Results are summarized in Figure 4. Observers rated male
and female faces differently, and their ability to do this
depended on the type of motion information available.
Analysis of variance confirmed this pattern of results by
showing a significant interaction between the type of mo-
tion and the sex of the face [F(2,45) 3.4, p .05]. There
were simple main effects of sex for combined [F(1,45)
40.0, p .05], facial alone [F(1,45) 26.9, p .05]
and head alone [F(1,45) 7.5, p .05] conditions, and
these results show that all types of motion contained use-
ful cues to sex. One-sample t tests comparing ratings to
the theoretically neutral rating value of 3.5 showed that
stimuli with combined motion were rated significantly
884 Current Biology Vol 11 No 11
Figure 5
Results of the 2-AFC sex judgement task. Observers (N 14) saw
pairs of animations, one male and one female, and had to indicate
which was which. Performance was significantly better than chance
would have allowed only when stimuli were played normally (upright
and forward), although inverted faces were also categorized marginally
better than chance would have allowed. The detrimental effect of
playing animations backward highlights the importance of the pattern
of movement for sex judgements as opposed to the low-level motion
or static cues that remain the same when animations are played
backward. Error bars show standard errors.
differently from neutral for both male [t(15) 5.7, p
.05] and female [t(15) 3.4, p .05] items, as were
stimuli containing only nonrigid facial motion [male:
t(15) 5.9, p .05; female: t(15) 3.4, p .05]. With
rigid head motion alone, male stimuli items were rated
as marginally differentl from neutral [t(15) 1.9, p
.07], but female items were not rated as significantly dif-
ferent from neutral (p .1). For this task, nonrigid motion
produced better performance than rigid motion, and this
pattern is opposite to that found for the identity-sorting
task. This suggests that the differences reported between
the two types of information are not simply a function of
any limitation in our animation of nonrigid motion. The
results contrast with previous evidence from experiments
in which point light stimuli were used. In these experi-
ments, nonrigid motions were found to be more useful
for sex judgements, with no difference for identity judge-
ments [3]. However, in this previous work rigid motions
were posed nods, shakes, and rocks of the head, not natu-
rally occurring movements, and they would have provided
clues to underlying differences in shape not available
here. Performance in the previous study was also above
chance and at a similar level to that reported here,
61.9% [3].
We further investigated sex judgements by using a two-
alternative forced-choice task (2-AFC), a task that avoids
any bias associated with the shape and texture of the
average head. Fourteen observers who had not taken part
in any of the previous experiments were presented with
pairs of stimuli, one male and one female, and had to
decide which was which. Stimuli were presented in two
blocks, with upright and inverted stimulus pairs randomly
interleaved in one block and forward and backward stimu-
lus pairs randomly interleaved in the other. The order of
the blocks was balanced, and both stimuli in a pair were
always shown in the same condition. All stimuli contained
both rigid head and nonrigid facial movements.
Results are summarized in Figure 5, with the percentage
of correct categorizations for normal stimuli collapsed
across both blocks. Paired t tests showed no differences
between presentation conditions. However, one-sample
t tests showed that only for stimulus pairs presented nor-
mally was performance significantly above chance (50%);
t(13) 5.6, p .05 and t(13) 3.2, p .05 in the
blocks with inverted and backward stimuli, respectively.
Performance for inverted stimuli was marginally above
chance, with t(13) 1.9 and p .1. Levels of performance
were not high with these stimuli because most of the
normal spatial cues to sex, including color and shape [18],
are kept constant. Static or low-level motion cues alone
cannot explain the pattern of performance observed, as
these remained the same between presentation condi-
tions. Instead, there appear to be direction-specific pat-
terns of movement for upright faces that differentiate
between male and female.
Conclusions
The results show that both rigid head and nonrigid facial
movements provide useful information for categorizing
both sex and identity. There are differences in the ways
that people move their heads and their faces, and we can
recover and use these identity cues. Rigid head move-
ments appear to be particularly useful for distinguishing
between individuals, and nonrigid motion appears to be
useful for categorizing on the basis of sex. This may be
because rigid head movements can be idiosyncratic, while
most nonrigid facial motion is functionally related to spe-
cific aspects of speech and expression, which have anyway
to be processed independently from identity [19]. Effects
of inversion show that low-level motion cues are not suffi-
cient to explain performance and suggest that dynamic
information is encoded by a model-based system special-
ized for upright faces. Backward movement, although dis-
criminable, disrupts sex judgements, and this result shows
that the direction of patterns of movement can be critical.
Supplementary material
Examples of animations used as stimuli are available on the internet at
http://images.cellpress.com/supmat/supmatin.htm.
Acknowledgements
This work was supported by the the Engineering and Physical Sciences
Research Council (MS2715). Thanks go to Peter McOwan, Szonya Durant,
Colin Clifford, and Branka Spehar for comments on drafts of the paper and
to Glyn Cowe for technical assistance.
 Brief Communication 885

References
1. Basilli J: Facial motion in the perception of faces and emotional
expression. J Exp Psychol 1978, 4:373-379.
2. Basilli J: Emotion recognition: The role of facial movement and
the relative importance of upper and lower areas of the
face. J Pers Soc Psychol 1979, 37:2049-2058.
3. Bruce V, Valentine T: When a nods as good as a wink: The role
of dynamic information in facial recognition. In: Practical
aspects of memory: Current research and issues, vol. 1. Edited by
Gruneberg MM, Morris PE, Sykes RN. Chichester, UK: Wiley;
1988:169-174.
4. Christie F, Bruce V: The role of dynamic information in the
recognition of unfamiliar faces. Memory and Cognition 1988,
26:780-790.
5. Knight B, Johnston A: The role of movement in face recognition.
Visual Cognition 1997, 4:265-273.
6. Lander K, Christie F, Bruce V: The role of movement in the
recognition of famous faces. Memory and Cognition 1999,
27:974-985.
7. Lander K, Bruce V: Recognizing famous faces: exploring
the benefits of facial motion. Ecological Psychology 2000,
12:259-272.
8. Pike GE, Kemp RI, Towell NA, Phillips KC: Recognizing moving
faces: the relative contribution of motion and perspective
view information. Visual Cognition 1997, 4:409-437.
9. Vetter T, Troje N: Separation of texture and shape in images
of faces for image coding and synthesis. J Opt Soc Amer
1997, 14:2152-2161.
10. Smith AR: Digital humans wait in the wings. Scientific American
2000, 283:72-78.
11. Johannsson G: Visual perception of biological motion and a
model for its analysis. Percept Psychophys 1973, 14:201-211.
12. Johannsson G: Studies of perception of locomotion. Perception
1977, 6:365-376.
13. Ullman S: The interpretation of visual motion. Cambridge,
Massachusetts: MIT Press; 1979.
14. Stone JV: Object recognition using spatiotemporal signatures.
Vision Res 1998, 38:947-951.
15. Marr D, Nishihara HK: Representation and recognition of three-
dimensional shapes. Proc R Soc Lond B Biol Sci 1978,
200:269-294.
16. Efron B: Computer-intensive methods in statistics. Scientific
American 1983, 248:116-130.
17. Valentine T: Upside-down faces: a review of the effect of
inversion upon face recognition. Br J Psychol 1988, 79:471-
491.
18. Hill H, Bruce V, Akamatsu S: Perceiving the sex and race of
faces: the role of shape and colour. Proc R Soc Lond B Biol
Sci 1995, 261:367-373.
19. Bruce V, Young A: Understanding face recognition. Br J Psychol
1986, 77:305-327.